Welcome to a forest in Provence, France. In these IPython notebooks you'll explore classifying the birds around you from audio clips of their songs and calls. The code is available on Github
The data used in this activity were provided during the NIPS4B workshop in Nevada 2013, in the Bird challenge, organized by Bas, Dufour and Glotin. They shall not be used for another purpose. Publication or report using them shall cite:
Bas Y., Dufour O., Glotin H., (2013) 'NIP4B Bird Challenge Overview', in Proc. of 'Neural Information Processing Scaled for Bioacoustics: from Neurons to Big Data- NIP4B', Ed. Glotin et al., joint to NIPS Conf., ISSN 979-10-90821-04-0, http://sabiod.org
In [1]:
import pandas as pd
import numpy as np
import bird_data as data
import wav_utils
from IPython.core.display import HTML
# Plot in the notebook
%pylab inline
In [2]:
data.call_df.head(10)
Out[2]:
In [3]:
HTML(data.label_df.head(10).to_html())
Out[3]:
In [4]:
data.call_df.type.unique()
Out[4]:
In [4]:
In [5]:
wavs = data.get_wav_dict()
In [6]:
three_call_clipnames = data.label_df.index[data.label_df.sum(axis=1)==3]
In [7]:
plot(wavs[three_call_clipnames[1]])
Out[7]:
In [8]:
wav_utils.view_clip(three_call_clipnames[4])
Out[8]:
In [9]:
wav_utils.view_clip(three_call_clipnames[1])
Out[9]:
In [10]:
wav_utils.view_clip(data.label_df.index[32])
Out[10]:
In [11]:
wav_utils.view_clip(data.label_df.index[8])
Out[11]:
In [12]:
wav_utils.view_clip(data.label_df.index[31])
Out[12]:
In [13]:
wav_utils.view_clip(data.label_df.index[34])
Out[13]:
In [13]: